common-session-retrospective
Analyze conversation corrections to detect skill gaps and auto-improve the skills library. Use after any session with user corrections, rework, or retrospective requests. After finding correction loops, also load +common/common-learning-log to persist mistake entries to AGENTS_LEARNING.md.
70
87%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Passed
No known issues
Quality
Discovery
89%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is a strong description that clearly communicates both what the skill does and when to use it, with explicit trigger conditions. The trigger terms are natural and well-chosen for the meta-learning domain. The main weakness is that the specific actions (especially 'auto-improve the skills library') could be more concrete about what changes are actually made.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names the domain (conversation corrections, skill gaps) and some actions (analyze, detect, auto-improve, persist mistake entries), but the actions are somewhat abstract—'auto-improve the skills library' is vague about what concrete steps are taken. | 2 / 3 |
Completeness | Clearly answers both what ('Analyze conversation corrections to detect skill gaps and auto-improve the skills library') and when ('Use after any session with user corrections, rework, or retrospective requests'), with explicit trigger guidance and even a follow-up action instruction. | 3 / 3 |
Trigger Term Quality | Includes natural trigger terms users or the system would use: 'corrections', 'rework', 'retrospective', 'skill gaps', 'correction loops', 'mistake entries'. These cover the likely vocabulary around post-session improvement workflows. | 3 / 3 |
Distinctiveness Conflict Risk | This is a highly specific niche—analyzing corrections to improve a skills library and persisting to AGENTS_LEARNING.md. It is unlikely to conflict with other skills due to its unique focus on meta-learning and self-improvement workflows. | 3 / 3 |
Total | 11 / 12 Passed |
Implementation
85%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a well-structured, concise skill that clearly defines a multi-step retrospective workflow with appropriate progressive disclosure to reference files. Its main weakness is the lack of concrete, inline examples—such as a sample correction signal, a sample classification, or a sample report output—which would make the protocol more immediately actionable without needing to consult the reference file. The bundle files (methodology.md) were not provided, making it impossible to verify that the deferred content fills these gaps.
Suggestions
Add a brief inline example of one correction loop detection and its resulting fix proposal to make the protocol more immediately actionable without requiring the reference file.
Include a minimal example of the report output format (step 7) directly in SKILL.md so Claude knows the expected output shape without loading methodology.md.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Every line serves a purpose. No unnecessary explanations of what retrospectives are or how corrections work. The protocol steps, guidelines, and anti-patterns are all lean and assume Claude's competence. | 3 / 3 |
Actionability | The 7-step protocol provides a clear process, but guidance remains at the procedural level without concrete examples of what a correction signal looks like, what a fix looks like, or what the report output format should be. Key details like the trigger miss schema and report template are deferred to references/methodology.md which is not provided in the bundle. | 2 / 3 |
Workflow Clarity | The 7-step workflow is clearly sequenced with logical progression from Extract → Classify → Check → Propose → Implement → Log → Report. The Trigger Miss Check (step 3) acts as a validation checkpoint, and the logging step ensures persistence. The workflow is well-structured for a process that doesn't involve destructive operations. | 3 / 3 |
Progressive Disclosure | Clean structure with SKILL.md as a concise overview and heavier content (signal tables, taxonomy, report template, trigger miss schema) appropriately deferred to references/methodology.md. The file tree at the top makes navigation clear, and references are one level deep with specific anchor links. | 3 / 3 |
Total | 11 / 12 Passed |
Validation
81%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 9 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
metadata_version | 'metadata.version' is missing | Warning |
metadata_field | 'metadata' should map string keys to string values | Warning |
Total | 9 / 11 Passed | |
- Repository
- HoangNguyen0403/agent-skills-standard
- Commit
3df717f
- Reviewed
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.